Dental Eruption and Growth in Hyracoidea (Mammalia, Afrotheria) ROBERT J. ASHER,*1 GREGG F. GUNNELL,2 ERIK R. SEIFFERT,3 DAVID PATTINSON,1 RODOLPHE TABUCE,4 LIONEL HAUTIER,*4 HESHAM M. SALLAM*5,6 1, Department of Zoology, University of Cambridge, CB2 3EJ, UK; 2, Duke Lemur Center, Division of Fossil Primates, Durham NC 27705, USA; 3, Department of Integrative Anatomical Sciences, University of Southern California, Los Angeles CA 90033, USA; 4, Institut des Sciences de l’Évolution (UM, CNRS, IRD, EPHE), c.c. 064, Université de Montpellier, Place Eugène Bataillon, 34095 Montpellier cedex 05, France; 5, Mansoura University Vertebrate Paleontology Center, Department of Geology, Mansoura University, Mansoura, 35516, Egypt; 6, Department of Evolutionary Anthropology, Duke University, Durham, NC 27708, USA. * corresponding authors: r.asher-at-zoo.cam.ac.uk, Lionel.Hautier-at-univ-montp2.fr, [email protected] 1 ABSTRACT---We investigated dental homologies, development, and growth in living and fossil hyracoids, and tested if hyracoids and other mammals show correlations between eruption patterns, gestation time, and age at maturity. Unlike living species, fossil hyracoids simultaneously possess replaced P1 and canine teeth. Fossil species also have shorter crowns, an upper and lower I3 locus, an upper I2, and a hypoconulid on m3. Prenatal specimens of the living Procavia capensis and Heterohyrax brucei show up to three tooth buds posterior to upper dI1 and anterior to the seven upper cheek teeth that consistently erupt; these include an anterior premolar but not a canine. Most lower cheek teeth finish eruption during growth in hyracoids, not after growth as in most other afrotherians. All hyracoids show the m1 at (lower) or near (upper) the beginning of eruption of permanent teeth; M3/m3 is the last permanent tooth to erupt. The living P. capensis erupts most lower antemolar loci before m2. In contrast, fossil hyraxes erupt lower antemolars after m2. While the early eruption of antemolars correlates with increased gestation time and age at maturity in primates and Tupaia (i.e., "Schultz's Rule"), and while modern hyraxes resemble some anthropoid primates in exhibiting long gestation and eruption of antemolars at or before molars, eruption patterns do not significantly co-vary with either life history parameter among afrotherians sampled so far. However, we do observe a shift in eruption timing and crown height in Procavia relative to fossil hyracoids, mirroring observations recently made for other ungulate-grade mammals. 2 PLAIN LANGUAGE SUMMARY---Our work examines how teeth erupt in living and fossil hyraxes (also known as dassies) and how they differ from their extinct relatives and other mammals. Fossil hyraxes differ from living ones in showing baby teeth for the first premolar and canine, and in having shorter tooth crowns. Hyrax fetuses help us to identify the first molar-like tooth that consistently erupts as a premolar, not a canine. Most of the permanent teeth erupt and are in use while hyraxes are growing, not after they reach adult size as in hyrax relatives like many tenrecs, golden moles, sengis, and manatees. In hyraxes the first molar is generally the first permanent tooth to erupt, and the last molar erupts after all of the others. Living hyraxes have fewer teeth in the front of the jaws and show eruption of several other permanent teeth prior to the second molar. In contrast, fossil hyraxes erupt most of their permanent teeth after their second molar. The eruption of some teeth with or before the molars correlates with a longer pregnancy time and age at maturity in primates, and living hyraxes resemble the eruption patterns and long pregnancies seen in some monkeys and humans. Mammals more closely related to hyraxes do not show such correlations, but eruption timing does correlate with increased crown height in living hyraxes, an association recently made by other investigators for certain fossil rhinoceroces and extinct mammals native to South America. 3 INTRODUCTION Living hyraxes, or dassies (Procaviidae, Hyracoidea, Afrotheria), consist of three genera and four species: Procavia capensis, Heterohyrax brucei, Dendrohyrax arboreus, and Dendrohyrax dorsalis (Shoshani 2005). Adults average between 2- 4kg; males of P. capensis can approach 5kg in body mass. All are semi- to very gregarious herbivores found in arid and forested habitats. They are almost entirely African in their extant distribution. P. capensis is the most widespread, from South Africa into the Arabian peninsula and Levant (Fischer 1992; Nowak 1999). Hyracoids were much more diverse in the geological past, including forms convergent upon antelopes (Antilohyrax), rhinos (Titanohyrax), suids (Geniohyus), and a late Eocene/early Oligocene genus with a pneumatized, sexually dimorphic mandibular resonating chamber (Thyrohyrax; see Schwartz et al. 1995; Rasmussen & Simons 2000; DeBlieux et al. 2006). Some aspects of their anatomy are similar to perissodactyls (Fischer 1992), but they are not closely related to them, beyond their status as placental mammals. Hyracoids belong in Afrotheria along with proboscideans, sirenians, Orycteropus, macroscelidids, tenrecids, and chrysochlorids (Murphy et al. 2001; Asher & Seiffert 2010; Tarver et al. 2016). Hyracoids have a good fossil record, including postnatal growth series of two genera known from near the Eocene-Oligocene boundary in Egypt: Thyrohyrax and Saghatherium. Ongoing paleontological work over several decades (Bown et al. 1982, 1988; Rasmussen & Simons 1988; Gheerbrant et al. 2007; Tabuce et al. 2007; Seiffert 2006, 2007; Rasmussen & Gutierrez 2010; Barrow et al. 2010, 2012; Tabuce 2016) has uncovered thousands of individual hyracoid fossils, and these two taxa are among the most common fossils recovered at Paleogene sites in northern Africa. While the vast majority are fragmentary, several are more complete and represent a 4 number of distinct ontogenetic stages. Here, we take advantage of this collection in order to investigate their growth and dental homologies. We combine our observations on these fossils with data on growth and dental eruption among living hyraxes and other taxa in order to investigate the pattern and process of dental eruption in mammals. Dental ontogeny in Procavia capensis Studies of dental eruption in the extant rock hyrax (Procavia capensis) are based on populations from the Levant (Roche 1978) and South Africa (Fairall 1980; Fourie 1983; Steyn & Hanks 1983; Fisher & Parkington 2014), as summarized in Table 1. These authors assumed a full dental formula of 1.0.4.3/2.0.4.3 (incisors.canines.premolars.molars; see below and Fig. 1). According to Fourie (1983; see also Fischer 1992: table 62), the earliest, approximate chronological age of hyraxes when the lower dp4, m1, m2, and m3 finish erupting is (respectively) 5, 13, 21, and 49 months post-birth. Upper teeth tend to erupt after their lower counterparts, and the upper M3 finishes erupting around 60 months post-birth, notably comprising about half of this animal's typical 10-12 year lifespan. Roche's (1978) eruption data from the upper toothrow are mostly consistent with Fourie (1983), with the exception of the upper premolars. According to Roche (1978), upper premolars erupt after M1 but before M2, whereas Fourie (1983: table 17) implies that they erupt prior to M1. Fairall (1980: table 4) gave eruption data for a captive South African population of P. capensis, fed a mixture of alfalfa and commercial rabbit feed. He reported the lower m1 as the first to erupt, followed by i1-2, p1-2, m2, then p3, then m3 in the dentary. Strangely, he depicted the lower p4 locus as absent ("no tooth" in his table 4) throughout growth; to our knowledge no other author has ever 5 claimed that P. capensis lacks a lower p4, nor have we seen any specimen that lacks one. In the upper dentition, he reported M1 as the first to erupt, followed by I1 and P1-2, then P3-4, then M2, then M3. This eruption sequence is generally consistent with that of Fourie (1983), except that in the lower dentition p3 and m2 are switched, and that Fairall (1980) neglects the p4 locus. Fairall's (1980) sequence for the upper dentition shows an earlier eruption for M1 (ahead of I1 and the premolars) compared to Fourie (1983). Even more discrepant are Fairall's (1980) much earlier timings for eruption at the m1 and m3 loci, which he claimed to have observed in the lower dentition at (respectively) two and 24 months. However, Fairall (1980:21) noted that "criteria for a fully erupted tooth were different" and that "it is possible that M3 was not fully erupted at 36 months" (compared to over 60 months reported by Roche, 1978), without further elaboration. Despite these discrepancies, and for wild populations, Fourie (1983: 97) wrote that "provided the date of collection within the study area is known, hyrax skulls up to the age of 36 months can be aged with a high degree of accuracy (to within ± 1 month) using eruption and replacement criteria." Although Fairall (1980) and Fourie (1983) did not report identical values for ages at molar eruption, they did report mutually consistent patterns of skull growth. Fairall (1980: table 2) reports cranial metrics for hyraxes of known ages between 1 day and 36 months; Fourie (1983: table 22) provided similar data for hyraxes between 12 and 108 months, as depicted in supplementary Fig. S1. Where the ages overlap in the two studies (one to three years), reported jaw lengths also overlap, growing from ca. 60mm at year one to just over 70mm at year three, when both male and female hyraxes greatly slow their rates of growth (Fig. S1). Moreover, the skull metric-age regression equations given by Fourie (1983: table 23) provide estimates 6 of age derived from jaw lengths that (as expected) fit reasonably on the curve that represents actual ages. Hyracoid dental homologies The authors above treated the anterior-most permanent tooth typically found in the maxilla of fully grown Procavia as the P1, not canine.